New Ways to Monitor Diabetes

05 February 2012

Interview with

Charles Rennie, University of Bristol

Ben -   One of the well-known consequences of being overweight, especially for a long time, is an increased risk of developing diabetes.  This leaves patients unable to control their own blood glucose level and causes serious complications including kidney failure and loss of sight.  Synthetic chemists like Bristol University's Charlie Rennie are leading the charge to find useful new molecules...

Charles -   The World Health Organisation has estimated that 346 million people worldwide are currently suffering from diabetes, so there's a massive demand for bringing new medicines, new treatments to market to help these people.  So, one of the main focuses of our research at the moment is, can we find a way in which we can offer continuous glucose monitoring.  So a way of continuously monitoring sugar levels within diabetic patients.

Ben -   At the moment, diabetics tend to monitor their own blood glucose level by a very old method of pricking their finger and sampling the blood.  What's wrong with that technique?

Charles -   One of the main problems is that it relies on the patient themselves having to determine when they should measure their glucose levels.  They may forget at one point, or they may be busy and or not do it at the necessary time and therefore, they don't have an accurate reading of what their sugar levels are all the time.  This can have quite serious effects.  It can obviously lead to having quite high glucose levels when they feel that they don't have.

Ben -   So, they only monitor their blood glucose level when they remember, they then take an appropriate amount of insulin.  That might mean that they Insulin pumpare overmedicating themselves or medicating too often, or in fact, not medicating enough.  So what we need is a method where we have a constant idea of how much medication you need and we always give exactly the right amount of medication?

Charles -   Yeah, I feel that that's obviously the goal and where we'd like to be with the treatment of diabetes.  Currently, there are systems which do offer continuous glucose monitoring.  However, there's quite a few problems with these current systems.  Many of them are enzyme based and therefore have quite a limited lifetime.  They normally have a lifetime of up to about 3 days, so they can be implanted then have to be removed after 3 days.  Therefore, this has obviously not taken widespread use, as it's quite an awkward thing for the patient in order to have this implanted and then removed. 

They also tend to measure sugar levels from interstitial fluid, which is the fluid which surrounds cells in the body, and therefore, they're not in direct contact with the bloodstream.  There seems to be a lag between the actual glucose levels in the blood and the readout you normally get from these devices.  So, obviously there are still quite a few improvements that can be achieved in this and this is probably where our research comes in.  We have looked at making ways in which continuous glucose monitors can be used 24/7 for prolonged periods of time and offer a really accurate readout of glucose levels for diabetic patients and therefore improve the treatment and management of the disease.

Ben -   There are obviously lots of challenges in getting that right.  What's the approach that you're taking?

Charles -   The approach we're taking at the moment is; can we really focus on trying to bind glucose initially?  Can we find a system which selectively binds glucose from an aqueous media such as water?  Therefore, we're looking to take it right back and design molecules initially based on molecular modelling, matching up the different parts of the sugar molecule to a receptor in a cage structure around that, and then using chemical synthesis to put this molecule together, in order to give a highly selective molecule for binding sugars.  Also, we'd like a compound which would give an output upon binding, such as a change in optical properties, or a changing voltage or something which we then could relate back to say, how much sugar was bound at any particular moment.

Ben -   So you're looking at the molecular structure, both of the glucose - the thing you want to bind - and then of candidate molecules that would lock very tightly to that glucose and then give you some kind of readout.  So what are the sorts of candidates you're looking at?

Charles -   The kind of candidates we're looking at are quite rigid structures and we want them to be positioned quite neatly and compactly around the sugar molecule.  Candidate molecules also have to have a certain amount of water solubility, so they have to have groups on them which will make them go into water.  This as the medium which we want to operate in.  We also want them to have the best matchup to the functionality on the glucose, on the sugar molecule.  So we're giving them dual characteristics, so there are dual properties in which certain parts of the molecules will entrap certain parts of the sugar molecule and then other parts will entrap the other parts of the sugar molecule to give a more closely knit binding.

Ben -   And how's it going so far?  Do you have a range of molecules that you're currently testing?

Charles -   Currently, we've managed to synthesise a couple of molecules which offer selective glucose binding from an aqueous medium.  They seem to be stable in blood plasma. So currently, we have quite a good starting point.  However, there's a few problems surrounding some of these.  The synthesis of these molecules is long winded, with quite a lot of steps so obviously, it's going to cost a lot of money in the long term, and obviously takes a lot of time.  We'd like to reduce the number of steps in order to make these compounds. 

Also, the changing properties, which I eluded to earlier, don't seem to be as strong and clear-cut as we would quite like at the moment, so we'd like a more distinct change upon binding of the sugar.  Therefore, there's quite a bit of work surrounding the tuning of these properties in order to get a better readout of how much sugar will be bound.

Ben -   If this is something where the ultimate aim is to put it inside somebody, we also need to be very careful that it isn't going to react to biological tissue and of course, is not going to provoke an immune response.

Charles -   Yes.  Initially, we'd just like to find a system which will bind sugar.  But in the long term, if we do find a system which seems to be adequate in doing this, then we then do a full range of screens and tests on these compounds in order to determine their biological activity, stability and toxicity.

We'll also look to consider how it would be best to administer this kind of compound into the body.  Would it be best to have it localised in a certain place, as an implant, or would it best to inject it into the system?  We're thinking at the moment that an implant would be the best situation as we can implant in direct contact with the blood flow, hopefully, in order to get an accurate readout of glucose levels, but also to prevent the compound from moving around the body and interfering with other areas.

Ben -   Once you have found your perfect compound that's biologically compatible, it locks in very specifically to glucose, and then gives you a very distinct change in properties, what do you then see us doing with that readout?

Charles -   Ideally, we'd like a system which the change in properties were to be able to be detected externally by wearing something such as a watch.  Then you could get a readout on the watch of your glucose levels, at any point you want.  You could have an alarm system which tells you when it's going high.  But also, we envision connecting our system up to something which currently exists, such as an insulin pump, which will then inject appropriate amounts of insulin into the patient.  Then we have set up a system almost like an artificial pancreas, and that would be a very strong way of treating diabetes and managing the current disease.

Add a comment

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.